UNIVERSITÀ
DEGLI
STUDI
DI
SASSARI
DOTTORATO IN RIPRODUZIONE,PRODUZIONE E BENESSERE ANIMALEXX CICLO 2004-2007
Coordinatore: Prof. Salvatore Naitana
THE RELATIVE EFFECTIVENESS OF
2-
METHOXYESTRADIOL ON UNDIFFERENTIATED AND DIFFERENTIATED CELLS OFGLIAL AND NEURONAL ORIGIN
DOCENTE GUIDA: TESI DI DOTTORATO DEL: PROF.VITTORIO FARINA DOTT.PAOLO MANCA
2
3 INTRODUCTION CYTOSKETETON General features 8 Microfilaments General features 9 Actin 9 Microfilaments assembly 9
Microfilaments organization and functions 10 Intermediate filaments
General features 11
Structure 12
Types of intermediate filaments 12
Types I and II (Acidic and Basic Keratins) 12
Type III 13
Type IV 13
Type V 14
Tubulin
General features 15
Microtubules and tubulin structure 16
Microtubules assembly 19
Microtubules dynamics in vivo and in vitro 22
Microtubules-stabilizing proteins 23
Post-translational modifications 25
Tyrosination/detyrosination: the tyrosination cycle 25
Acetylation/deacetylation 27
Polyglutamilation 28
Poliglycilation 28
Palmytoylation 29
Phosphorylation 29
STEROID HORMONES, NEUROSTEROIDS AND NEUROACTIVE STEROIDS
Steroids hormones 31 Neurosteroids 31 Neuroactive steroids 32 Steroid synthesis in the brain: conversion of cholesterol to pregnenolone 34 Steroid synthesis in the brain: conversion of androgens to estrogens 35
4
3β-hydroxysteroid dehydrogenase: inactivation of metabolite of steroid hormone 36 17β-hydroxysteroid dehydrogenase: regulation of biological activity of steroid hormones
36
2-METHOXYESTRADIOL: BIOCHEMIICAL FEATURES AND ITS ACTIVITY IN VITRO AND IN VIVO
General features 38
Biochemical and physiological properties 38
2-Methoxyestradiol affinity for oestrogen receptors 39 2-Methoxyestradiol has an antiangionetic activity: inhibition of HIF-1 alpha expression 40 2-Methoxyestradiol exerts an antimitotic activity: abnormal spindle, tubulin
depolymerization & alteration of tubulin dynamics
41 2-Methoxyestradiol alters cell mobility, cell adhesion and trans-well migration in
vivo and in vitro
43 2-Methoxyestradiol interacts with tubulin and inhibits colchicines binding site in
vivo and in vitro
44
2-Methoxyestradiol induces apoptosis 44
2-Methoxyestradiol modifies tubulin expression 46
2-METHOXYESTRADIOL EXPOSED GLIAL AND NEURONAL CELL LINES: LITERATURE REPORTS
2-Methoxyestradiol-exposed cells of glial origin 48 2-Methoxyestradiol-exposed cells of neuronal origin 50
CELL DIFFERENTIATION
51
AIMS OF THE STUDY 52
MATERIALS & METHODS
Cell culture and treatments 55
Microscopic examination and proliferation assay (MTT test) 56
Indirect immunofluorescence 57
Staining with Hoechst H33342/propidium iodide (PI): viability assay 58
Western Blot 58
Statistical analysis 59
RESULTS
UNDIFFERENTIATED C6&C1300 CELLS
5
Hoecst 33258/propidium iodide (PI) staining: viability test 62 MTT test: cell proliferation assay 63 Western blot
Total α-tubulin 64
Acetylated α-tubulin 64
Tyrosinated α-tubulin 65
TAPSIGARGIN-DIFFERENTIATED C6 & DB-CAMP-DIFFERENTIATED C1300
CELLS
PHACO: microscopic examination 66
Hoecst 33258/propidium iodide (PI) staining: viability test 66 MTT test: cell proliferation assay 67
Western blot analysis 67
Total α-tubulin 68 Acetylated α-tubulin 68 Tyrosinated α-tubulin 68 FIGURES UNDIFFERENTIATED C6 CELLS PHACO 71 Vitality test 72 MTT test 75
Western blot analysis
Total α-tubulin 76 Acetylated α-tubulin 78 Tyrosinated α-tubulin 82 UNDIFFERENTIATED C1300 CELLS PHACO 73 Vitality test 74 MTT test 75
Western blot analysis
Total α-tubulin 77
Acetylated α-tubulin 79
Tyrosinated α-tubulin 81
TAPSIGARGIN-DIFFERENTIATED C6 CELLS
PHACO 82
Vitality test 83
MTT test 86
Western blot analysis
Total α-tubulin 87
6 Acetylated α-tubulin 89 Tyrosinated α-tubulin 91
DB-CAMP-DIFFERENTIATED C1300 CELLS
PHACO 84
Vitality test 85
MTT test 86
Western blot analysis
Total α-tubulin 88 Acetylated α-tubulin 90 Tyrosinated α-tubulin 92 DISCUSSION 94 BIBLIOGRAFY 102 SITOGRAFY 118
7
8
CYTOSKELETON
General featuresCytoskeleton is considered to be the cellular scaffolding or skeleton. Cytoskeleton is contained, as all other organelles, within the cytoplasm of all eucaryotic cells (Fig. 1). Recent research carried out with electron microscopy on some
eubacteria has demonstrated that cytoskeleton can be present in prokaryotic cells too (Mayer, 2006). Cytoskeleton is a dynamic fibrillary structure that is essential to keep and determine any cellular shape changes. It is well known that cytoskeleton make possible some cell motion (by characteristic cellular structures such as flagella and
cilia), and plays a crucial role both in intra-cellular transport such as in the movement of vesicles, organelles and cellular division. Cytoskeleton is a bone-like structure floating around/within the cytoplasm essentially composed by three main filaments: microfilaments or actin filaments, intermediate filaments (IFs) and microtubules. Every cytoskeletal filaments are protein structures having the capability, except intermediate filaments, to rapidly polymerize and depolymerize. Accessory proteins bound the filaments and regulate their assembly.
Fig. 1. The eucaryotic cytoskeleton. Bovine pulmonary artery endothelial cells: actin filaments are in red (Texas Red-Phalloidin staining), microtubules are in green (Bodipy FL goat anti-mouse staining) and nuclei in blue (DAPI dye) (From http://rsb.info.nih.gov/ij).
9
MICROFILAMENTS
General features
Microfilaments are the thinnest filamentous cytoskeletal proteins, measuring approximately 7 nm in diameter (Fig. 2).
The single subunits of actin are known as globular actin (G-actin) while the whole filamentous polymer that is constitute of G-actin subunits is called F-actin.
Actin
G-actin is a globular structural, 42-47 kDa protein found in many eucaryotic organisms, with concentrations of over 100µM. It is also one of the most highly preserved proteins, differing by no more than 5% in species as diverse as algae and humans.
Microfilaments assembly
Actin is the monomeric subunit of microfilaments. Microfilaments have a polar structure like the microtubules since they have a fast growing plus end (also known as barbed end) and a slow growing
minus end (also known as pointed end). The terms barbed and
pointed come from the arrow-like appearance of the proteic structure as it has been seen in Electron Micrographs. Filaments elongate approximately 10 times faster at the plus end than at the minus end.
Fig. 2. Actin cytoskeleton. Mouse embryo fibroblasts: microfilaments are in green (FITC-phalloidin) (From http://rsb.info.nih.gov/ij).
10
When the polymerization rate at the plus end equals the depolymerization at the minus end the result is a tread-milling effect where the filaments move without changing its overall length.
Nucleation is the process of actin polymerization and it starts with the association of three G-actin monomers into a trimer. ATP-actin binds the plus (+) end then ATP (adenosine triphoshate) is hydrolyzed (half time 2 seconds) and the inorganic phosphate released (half time 6 minuts) (Pollard et al., 2004). ADP-actin dissociates from the minus end so that consequent increase in ADP stimulates the exchange of bound ADP-ATP leading to more ATP-actin units. The rapid turn-over is important for cell movement. Three proteins which are regulated by cell signaling mechanisms, work for microfilaments’ functionality. CapZ (known as end capping
proteins), prevents the addition or loss of monomers at the filaments when actin turn-over is unfavorable like in the muscle apparatus. Cofilin binds to ADP-actin units and promotes their dissociation from the minus end preventing their reassembly as well. On the other hand, profilin reverses cofilin effect by stimulating the exchange of bound ADP for ATP. Finally, the Arp2/3 complex nucleates new actin filaments while bound to existing filaments and create the branched network.
Microfilaments organization and functions
Microfilaments are an ubiquitary contractor organ not only involved in some cell-to-cell or cell-to-matrix junctions and transduction of signals but also engaged in keeping cellular shape or forming cytoplasmatic protuberances such as pseudopodia and microvilli. Moreover, microfilaments are crucial for cell movement (cytokinesis) and, along with myosin, muscular contraction.
11
Actin filaments are gathered in two types of structures: network and bundles with the characteristic features of a double-stranded helix. Networks are constituted by actin along with many actin-binding proteins (such as the Arp2/3 complex and filamin) that are characteristically concentrated at the cortical regions of the cells. Recently, it has been proved that actin network is involved in some cellular defensive mechanisms serving as barriers for molecular diffusion within the plasmatic membrane. In non-muscle actin bundles, the filaments are held together: they are kept parallel each other by actin-bundling proteins and/or cationic species. Bundles play a role in many processes such cytokinesis and cell movement.
INTERMEDIATE FILAMENTS General features
Intermediate filaments (IFs) are heterogeneous cytoskeletal structures formed by members of a family of related proteins (Fig. 3). IFs have a diameter between that of microfilaments and microtubules (7-11nm); IFs differ from microfilaments and microtubules not only for their chemical properties but also because they are very stable structures, formed by fibrotic subunit proteins. Most types of IFs are located in the cytosol between the nuclear envelope and the cell surface membrane.
Fig. 3. Intermediate filaments in epithelial cells (MDCK). Keratin filaments which are in red (Texas Red-Phalloidin staining), are concentrated around the edge of the cells, whereas DNA is visualized in green (Methyl green staining) (From http://en.wikipedia.org).
12
Structure
The domain structure of IFs molecules is conserved. Proteins have a typical non-α-helical (globular) domain at the N- and C-termini which surrounds the α-helical rod domain. The essential IFs building block is a parallel, in register dimer that is constituted through the interaction of the rod domain forming a coiled tail. Cytoplasmatic IFs assemble into non-polar unit-length filaments that then gather into longer structures. The characteristic anti-parallel orientation of the fibrotic subunits means that, unlike microtubules and microfilaments have a plus and a minus end, IFs lack any polarity and do not undergo tread-milling. New evidences have showed that IFs can be also dynamic and motile elements of the cytoskeleton (Helfand et al., 2004).
Types of intermediate filaments
There are about 70 different genes encoding for various intermediate filament proteins that are subcategorized into six types based on similarities in amino acidic sequence and protein structure.
Types I and II (Acidic and Basic Keratins)
They are an heterogenous class of IFs which constitute type I (acidic) and type II (basic) IFs proteins. They are further divided in two groups:
• Epithelial keratins (about 20): typical in epithelial cells; • Trichocytes keratins (about 13): hair keratins which make
up hair, nails, horns and reptilian scales. Acidic and basic keratins bind each other forming acidic-basic heterodimers that associate to form a keratin filament.
13
Types III
Four proteins which may form homo- or heteropolymeric proteins are classified as type III:
• Desmin: a structural component of the sarcomeres in muscle cells;
• GFAP (glial fibrillary, acidic protein): found in astrocytes and glia cells;
• Peripherin: found in periheral nerve;
• Vimentin: the most widely diffused of all IFs proteins which can be found in fibroblasts, leukocytes, blood vessels and endothelial cells. Vimentin provides for the cellular membrane and keeps some organelles in a fixed place within the cytoplasm.
Type IV
IFs that are exclusively present in the nervous system:
• Internexin: the major component of the IFs network in small interneurons and cerebellar granule cells;
• Nestin: expressed mostly in the nervous cells. They are implicated in the radial growth of the axon;
• Neurofilaments: found in high concentrations along the axon of vertebrate neurons;
• Synemin: originally found at the Z-band of avian and rodent skeletal and cardiac muscles. It occurs together with nestin and vimentin in glial progenitors during the early differentiation of the mouse central nervous system;
• Syncoilin: found at the neuromuscular junction, sarcolemma, and Z-lines.
14
Type V
Nuclear lamins.
15
TUBULIN
General features
Microtubules are long filamentous, tube-shaped protein polymers present in all eukaryotic cells. Microtubules are crucial in some cellular activities such as morphogenesis, cell mobility, cell signaling and organelles transport. In details, microtubules rearrange during cell division to form the mitotic spindle, which is fundamental not only for the segregation and the replication of sister cromatids, but also for the orientation of the cell cleavage plane. Moreover, microtubules are the most prominent constituent of the composite and extremely organized axonemal structure of
cilia and flagella. In addition, microtubules contribute in the generation of cell polarity and they work, with several motor proteins, as tracks along which organelles and/or vesicles are transported through the cell.
The great variety of microtubule functions comes from the amazing structural flexibility of microtubule organization depending, in a large extent, on their remarkable biochemical and functional properties. Microtubular arrays in eukaryotic cells are dynamic, able to assembly (polymerize), disassembly (depolymerize) and rearrange within few seconds or minutes. This typical property is called microtubular dynamics. It is well known that microtubular dynamics are based on intrinsic dynamic properties of the tubulin polymers themselves which are regulated by the biochemical features of the main microtubule building block, the tubulin α-β heterodimer.
16
Among tubulin biochemical attitudes, there is the tubulin intrinsic capability to interact in vivo and in vitro with a great number of other proteins such as small molecules, nucleotides or drugs that can modify the physiological characteristic of the whole polymer. Finally, it is of interest considering that tubulin is a GTPase as well. The phenomenon of GTP hydrolysis occurs during tubulin polymerization and is essential for the microtubular functional properties.
Microtubules and tubulin structure
Microtubules are composed of α-tubulin and β-tubulin heterodimers. α and β tubulin monomers are proteins of about 450 amino acids and each monomers have a molecular mass of about 50,000Da (Fig. 4). Tubulins are ubiquitous and, in eukaryotic organisms, several genes are known encoding for six α-tubulin and seven β-tubulin isotypes. Three α-tubulin isotypes (1, 2 and 4) and five β-tubulins (I, II, III, IVa and IVb) are expressed in the brain (Luduena et al., 1998; Lee et al., 1990; Redeker et al., 1998) (Table 1).
Fig. 4. Three-dimensional model of tubulin. The protein is a dimer consisting of two monomers that are almost identical in structure. Each monomer is formed by a core of two beta sheets (blue and green) surrounded by helices, and each binds to a guanine nucleotide (pink) (From http://www.lbl.gov).
17
Table 1. Overview of the various tubulin isoforms (from Bianchi et al., 2005).
Tubulins α and β isotypes
Comments References
α1/β2
• Highly expressed during brain development
• Associated with neurite outgrowth Luduena et al., 1998 α4 • Without a C-terminal tyrosine residue • Glutamylated at two residues Redeker et al., 1998
βI/βΙΙ • Highly expressed during brain development
Luduena et al., 1998
βIII • Neuronal specific Lee et al., 1990
βIVa, b • Highly expressed during brain development
18
α-tubulin and β-tubulin heterodimers arrange linearly into protofilaments that associate laterally forming microtubular structures with an outer diameter of 25 nm (Fig. 5).
Fig. 5. α-β tubulin heterodimers arrange linearly into protofilaments (From Westermann and Weber, 2003).
Different experimental conditions in vitro may vary the number of protofilaments of microtubules between 10 and 15, whereas they are 13 in physiological conditions in vivo. The organization of α- and α-β tubulin heterodimers in the microtubular structure is polarized, and this feature determines important structural and kinetic differences at the microtubule ends. α and β-tubulin heterodimers are arranged in a typical head-tail configuration in which they are polarized with a faster elongating end, named plus, exposing β subunits, and a slower elongating minus which show the alpha subunit in vivo. The minus end of the microtubules is associated with the centrosome and is localized close to the center of the cell, while the plus end is peripheral (Valiron et al., 2002) (Fig. 6).
19
Fig. 6. α and β-tubulin are arranged in a head-tail configuration in which they are polarized with a faster elongating end, named plus, exposing β subunits, and a slower elongating minus end showing the alpha subunit (From Alberts et al., 2002).
Microtubules assembly
Microtubule assembly proceeds in three phases: nucleation, elongation and steady-state.
Nucleation, also known as microtubules aggregation, is the phase of microtubular assembly in which a small cluster of α and β tubulin monomers aggregate to form a short microtubule nucleus arranged to initiate polymerization properly. The nucleation may give in a variety of conditions. Temperature is a fundamental factor; indeed, tubulin aggregation generally occurs at 30°-37°C, whereas lower temperature do not arise microtubules aggregation anymore. Another fundamental condition is microtubule incorporation as a complex with GTP whose energy input from hydrolysis works for tubulin dynamics. Nucleation is followed by the so-called elongation phase during which α-β dimers are added to the end of each microtubular structure just formed. Steady state is the last phase of tubulin assembly. In vitro studies have probed that a determined proportion of tubulin is in the polymerized
20
form, whereas another tubulin pool stands soluble (free proportion). Fascinatingly, the amount of soluble tubulin, which is in form of dimers during the steady state, varies with buffer conditions and this amount constitutes the critical concentration for microtubule assembly. Steady state is characterized by a constant GTP hydrolysis. Consumption of energy is necessary for maintenance of the steady state of microtubules since microtubules use energy to continuously provide molecular exchanges between α-β constitutive subunits and the tubulin molecules of the soluble pool (Margolis and Wilson, 1981). Several studies have clarified the mechanisms that are involved in such a continuous exchange. The first is known as tread-milling.
Tread-milling depends on a typical property of microtubular ends at the steady state (or pause): the plus end continuously incorporates new tubulin dimers, while at the same time the minus end looses them. The energy required for such a dynamic process comes from GTP hydrolysis but the precise mechanism that joins the functional asymmetry of the tubulin ends is still unknown. When tubulin ends are in a fixed position, tread-milling results in a sort of microtubular migration that reproduces the propagation of a wave of tubulin assembly and disassembly. By contrast, when microtubular ends are kept in a stationary position, tread-milling generates an apparent flux that travels from the minus to the plus end (Margolis and Wilson, 1981) (Fig. 7).
The second mechanism is known as dynamic instability. Some models of study showed that, while a population of microtubules exhibits a steady state, an individual microtubule persists in a slow phase of assembly and rapid phase of disassembly so that it never reaches an
21
equilibrium length. In details, the phase between slow assembly and fast disassembly is called catastrophe and the phase between disassembly and assembly is referred to as rescue (Fig. 7).
Taking into account these notions, microtubular dynamics are, by definitions, the intrinsic capability of microtubular polymers to rapidly change its biochemical and functional properties as consequence of overall environmental conditions. Polymerized microtubules are intrinsically labile but at the steady state, the whole microtubular system remains out of equilibrium. Events like temperature fall below 10°-15°C readily depolymerize microtubules. Moreover, other events such as removal of GTP, dilution of microtubule suspension with consequent decrease in soluble tubulin concentration, can induce tubulin disassembly. In addition, specific drug as colchicine, vinblastine or nocodazole can easily trigger tubulin depolymerization with different mechanisms of action. In details, colchicine and vinblastine perturb tubulin assembly into microtubules consequently inducing tubulin depolymerization. Naconazole sequesters tubulin dimers in inactive form that, decreasing the soluble pool of tubulin below the critical concentration,
22
induces microtubules depolymerization (Fig. 8).
Fig. 8. Tubulin pathways: perturbation of tubulin dynamics via specific MAPs such as Tau or traditional anticancer agents may lead to mitosis block and, by consequence, cell death (From http://merckbiosciences.com).
Microtubules dynamics in vivo and in vitro
In vivo studies on microtubule dynamics have revealed some extensive similarities with microtubule dynamics in vitro. Microtubule dynamics and dynamic instability (the main mechanism of microtubules turnover) has been clearly demonstrated in eukaryotic cells. However, differences in the rate of microtubular assembly (10-15% higher than in vitro microtubules formed by pure tubulin) and the control of the microtubular polymer by cell effectors has been found (Cassimeris, 1993). It is well known that cell regulation influences both microtubule nucleation and assembly. Nucleation in
vivo is only provided by the centrosome (also called microtubule-organizing center) and in most cells, the minus ends of microtubules are linked to the centrosome. Moreover, microtubule dynamics in vivo are regulated by many microtubule-associated proteins (MAPs) that
23
are usually classified in two main groups: proteins that stabilize and proteins that destabilize microtubules. Recent studies have also highlighted other function of MAPs depending on cellular context, especially with regard to regulation of the mitotic spindle, assembly, and overall microtubular organization and dynamicity in vivo. MAPs have been also discovered to interact with actin, intermediate filaments and to modulate microtubule dynamics as motor proteins.
Microtubules-stabilizing proteins
Microtubule-stabilizing proteins promote tubulin assembly and stabilize microtubules. Members of this family include the proteins tau, MAP2 (present in the axon and dendrites, respectively), and MAP4.
A new group of MAPs inducing much higher microtubular stability has been recently identified. This group includes Lis1(Lissencephaly Isolated Type 1), doublecortin, BPAG1 (Bullous Pemphigoid Antigen 1) and STOP (Stable Tubulin Only Peptide) proteins. Lis1 is distributed along microtubules, not only in neuronal cells but also in several cell phenotypes. Lis1 interacts with the microtubule motor dynein. In details, production of dynein increases retrograde movement of cytoplasmic dynein and leads to peripheral accumulation of microtubules. These findings suggest that the amount of Lis1 may stimulate specific dynein functions in neuronal migration and axon growth (Smith et al., 2000). Doublecortin is expressed in migrating and differentiating neurons and it can be associated in vitro and in vivo with microtubules acting as microtubular stabilizer. BPGA1 has the property to physically link actin, intermediate filaments and microtubular networks. BPGA1 can bind and stabilize microtubules in vitro so that microtubules become sensitive to various
24
depolymerizing agents, including cold. Cold stability can also be due to microtubules association with different variants of STOP that are typical calmodulin-regulated proteins. In spite of STOP come from a single gene, variants have different start genes of transcription and are expressed in many tissues at diverse stages of development. In neuronal cells, STOP proteins seem to be associated with microtubules: they are the major factors responsible for the slow turnover of neuronal microtubules and apparently required for neuronal differentiation. In addition, STOP proteins can be associated with microtubules in the mitotic spindle and such association appears to be fundamental for the progress of the mitotic process.
Though several proteins have been identified as microtubule-stabilizing proteins, their mechanisms of action are not well known yet. Op18/stathmin is a small ubiquitous protein which has been shown to destabilize microtubules by increasing the catastrophe frequency and regulating microtubules levels both in vivo and in vitro (Belmont and Mitchison, 1996). Two mechanisms involved in the Op18/stathmin mechanism of action have been postulated. Op18/stathmin has been hypothesized to sequester tubulins, lowering the total soluble amount of tubulin available for polimerization. Furthermore, Op18/stathmin may act directly on microtubules like a catastrophe-promoting factor. Another class of microtubule-stabilizing proteins is formed by the members belonging to the super-family of kinesin-related microtubules motor proteins. XKCM (Xenopus Kinesin-Related Protein) is a kinesin-related motor protein with the capability to inhibit mitotic spindle in vitro and to determine microtubule catastrophe in vivo. Heat-shock proteins (HSP27, HSP70 and HSP9) have been seen inhibiting microtubule formation in vitro at high concentrations as well. Finally, a small acidic polypeptide,
25
MINUS (Microtubule Nucleation Suppressor), biochemically purified from cultured cell lines and bovine brain, was ascertained to block microtubule nucleation in vitro in centrosome presence or absence.
Post-translational modifications
Microtubules heterogeneity is determined by several post-translational modifications such as tyrosination/detyrosination, acetylation, polyglycilation, polyglutamylation, phosphorylation and palmitoylation (Westermann and Weber, 2003). These modifications usually occur at the carboxy-terminal domain of the α-β heterodimer, in correspondence of the outer surface of the microtubule where it can interact with other proteins (Westermann and Weber, 2003).
Tyrosination/detyrosination: the tyrosination cycle
Tyrosination is a post-translational modification consisting in the addition of a carboxy-terminal tyrosine to most α-tubulins (Fig. 9). The first step of this modification is the removal of the carboxy-terminal tyrosine (detyrosination) by a tubulin tyrosine carboxipeptidase (TTCP). This post-translational modification occurs after microtubule assembly since the TTCP prefers polymers. (Argarana et al., 1978). The tyrosinated tubulin displays a carboxy-terminal glutamic acid that is usually referred as Glu-tubulin. In addition, losing Glu-tubulin the penultimate glutamate residue through the enzymatic action of a still unknown peptidase, ∆2 tubulin can be generated as well (Paturle-Lafanechère et al., 1991). Interestingly, ∆2 cannot be a substrate for the tubulin tyrosine ligase (TTL) and does not take part to the tyrosinated cycle (Paturle-Lafanechère et al., 1991). ∆2 represents the final stage of α-tubulin
26
maturation, it constitutes a considerable proportion of brain tubulin and it is known as a marker of very long-lived, stable microtubules (Paturle-Lafanechère et al., 1994).
An ATP-dependent reaction catalized by the activity of TTL can restore the carboxy-terminal tyrosine (Tyr-tub). TTL, in contrast with TTCP, prefers tubulin dimers as substrate so occurs exclusively before tubulin assembly (Argarana et al., 1977).
Fig. 9. The tyrosination cycle of tubulin. The carboxy-terminal tyrosine of α-tubulin can be removed by the α-tubulin tyrosine carboxypeptidase (TTCP) to generate Glu-tubulin (Glu-tub). In an ATP dependent reaction, the carboxy-terminal tyrosine (Tyr-tub) can be restored through the enzymatic activity of tubulin tyrosine ligase (TTL). Glu-tubulin can lose the penultimate glutamate residue through the activity of an unknown peptidase to generate ∆2 tubulin, which cannot function as substrate for TTL and is therefore removed from the cycle (From Westermann and Weber, 2003).
27
Although detyrosination itself does not stabilize microtubules, it is considered marker of dynamic microtubule. Moreover, detyrosination is thought to be crucial for the cross-talk of microtubule and intermediate filaments as well (Gurland et al., 1995) (Table 2). It has been shown that α-tubulin can incorporate 3-nitrotyrosine, a modified amino acid generated by the reaction of nitric oxide species with tyrosine. Interestingly, nitrotyrosination is catalyzed both in vivo and in vitro by the same enzyme, TTL, involved in tyrosination. It has been proposed that this incorporation is irreversible and that the accumulation of 3-nitrotyrosine leads to microtubule dysfunction and cellular injury (Zedda et al., 2004).
Acetylation/deacetylation
Acetylation is a well known post-translational modification that takes place at lysine-40 in the amino terminus of α-tubulin located in the lumen of microtubules (Nogales, 2001).
Even if the tubulin acetylatransferase enzyme has not been identified, recently the two enzymes catalyzing the reaction of deacetylation has been found. The enzymes are two histones deacetylase: HDAC6 (Histone Deacetylase 6) and SIRT2 (Sirtuins 2). HDAC6 belongs to the deacetylase family and it is mostly located in the cytoplasm in association with microtubules (Hubbert
et al., 2002). SIRT2 is the namesake of a family of closely related enzymes, the sirtuins, that are hypothesized to play a key role in the organism response to stresses such as heat or starvation (Frye et al., 2005).
28
Acetylation takes place after microtubule assembly. It is marker of stable microtubules being associated with stable microtubular structures such as axonemes (Sasse et al., 1988). Furthermore, has been proposed a role for tubulin acetylation in cell mobility on the basis of the fact that HADC6 over-expression increased the chemiotactic movements in cultured cell, whereas inhibition of the same enzyme blocks cell migration.
Polyglutamylation
Polyglutamilation is a not very usual form of reversible post-translational modification of glutamate residues occurring in α-β dimers and in two nucleosome assembly proteins (NAP1 and NAP2). For this reason, it seems that chromatin structural proteins are regulated by this post-translational modification.
In polyglutamylation, a polyglutamate side chain of variable length is attached, through an isopeptide bond, to the γ carboxy-group of a glutamate in the carboxy-terminal tail of tubulin (Boucher et al., 1994). Polyglutamylase and polydeglutamylase are the enzymes involved in glutamylation/deglutamylation processes.
Recent findings have revealed a prominent role of polyglutamylation in the interaction between microtubules and their associated proteins (i.e. MAPs 55) in centriole maturation/stability (Gagnon et al., 1996) and in ciliar/flagellar mobility as well (Million et al., 1999) (Table 2).
Polyglicylation
It is the most important post-translational modification of axonemal microtubule. Polyglicylation is the covalent attachment of a polyglycine side chain through an isopeptide bond to the carboxyl
29
group of conserved glutamate residues at the carboxyl-terminus of α-β tubulin. This modification was originally discovered in
Paramecium (Redeker et al., 1994) and later shown in mammalian neurons as well (Banerjee et al., 2002).
Presently, the enzymes catalyzing polyglycylation are not known. Genetic studies carried out on Tetrahymena thermophila have provided some information about the possible functions of polyglycylation. In Tetrahymena polyglycylation is essential for cellular activities related to axonemal organization, ciliary motility and cytokinesis (Xia et al., 2000) (Table 2).
Palmitoylation
Palmitoylation consists in the incorporation of the palmitic acid on the α-subunit at cysteine 376 (Caron et al., 1997). Studies carried out on the budding yeast Saccharomyces cerevisiae revealed that palmitoylation of α-tubulin is necessary for the correct positioning of astral microtubules and for the right microtubule interactions with the cell cortex (Caron et al., 2001) (Table 2).
Phosphorylation
Phosphorylation is not a common post-translational modification. It consists in the phosphorylation of a serine residue within the carboxy-terminal tail of β-tubulin. Phosphorylation plays a crucial role in the regulation of some MAPs and may be involved in neuronal differentiation since it regulates the axon outgrowth of class III ∆2-tubulin. (Eipper et al., 1974) (Table 2).
30
Table 2. Overview of the various tubulins modifications and their proposed functions (From Westermann and Weber, 2003).
Tubulin isotypes Post translational modifications Comments Proposed functions References α-β tubulins Tyrosination detyrosination Marker of dynamic microtubules • Cross-talk to intermediate filaments • Cell differentiation Cumming et al., 1984 Erck et al., 2005 α-tubulins Generation of ∆2 tubulin
Marker for stable very-longed microtubule • Removing tubulin from tyrosination cycle Paturle-Lafanechère et al., 1994 α-tubulin Acetylation deacetylation
Marker for stable microtubule • Regulation of cell mobility • Bindings of the MAPs Sasse and Gull, 1988
α-β tubulin Polyglutamylation Multiple glutamylation sites possible • Centriole maturation, flagellar motility • Regulation of interaction with MAPs Boucher et al., 1994
α-β tubulin Polyglicylation Multiple glycylation sites possible In Tetrahymena: • Ciliary motility • Cytokinesis Xia et al., 2000
α-tubulin Palmitoylation Present in budding yeast • Positioning of astral microtubules Caron et al., 2001
α-β tubulins Phosphorylation Mainly in β-tubulin • Neuronal differentiation
31
STEROID HORMONES, NEUROSTEROIDS AND
NEUROACTIVE STEROIDS Steroid hormones
Steroid hormones are mainly synthesized in the adrenal glands, the gonads and the feto-placental organ. They can easily cross the blood-brain barrier because of their high lipid solubility. The central and peripheral nervous systems are crucial target organs of steroid hormones. Moreover, an extensive steroid metabolism occurs in the brain, several brain regions and peripheral nervous system. Indeed, each of these areas are equipped with the enzymes necessary for the synthesis of steroid as well. The synthesis of steroids may take place particularly, but not exclusively, in myelinating glial cells, from cholesterol or steroidal precursors deriving from peripheral sources. Steroids have been found to play a pivotal role in the development, growth, maturation and differentiation of human brain. Indeed, the activity of steroidogenic enzymes have been identified in specific areas of human fetal brain.
Neurosteroids
The term neurosteroids was firstly defined by Baulieu and co-workers (Baulieu et al., 1990). They discovered that a number of steroid hormones existed in higher concentrations in the nervous system than in the plasma and that these steroids were synthesized in the brain. Nowadays it is known that neurosteroids can be metabolized in the brain from precursor compounds originating from endocrine sources, and can be synthesized de novo in the brain from cholesterol.
32
Neuroactive steroids are steroids that produce a rapid, non-genomic action in the brain, generally through the action on ligand - or voltage-gated channels (Stoffel-Wagner, 2001). For this reason neuroactive steroids differ significantly from classic steroid hormones since they mainly act upon intracellular receptors resulting in a long lasting, genomic effect (Truss et al., 1988). Neuroactive steroids can either positively or negatively modulate the activity of several ligand-gated ion channel-associated receptors including γ-aminobutyric acid type A (GABAA), serotonin type 3 (5-HT3),
glycine and glutamate N-methyl-D-aspartate (NMDAA),
α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) and kainite receptors.
Examples of neuroactive steroids are: a) the pregnane steroids 4-pregnene-3,20-dione (progesterone, P), 3β-OH-5-pregnen-20-one (pregnenolone, Preg), pregnenolone sulphate (PregS), 3α-OH-5β-pregnan-20-one (pregnanolone), 3α-hydroxy-5α-3α-OH-5β-pregnan-20-one (allopregnanolone, ALLO), 3α,21-dihydroxy-5α-pregnan-20-one (allotetrahydrodeoxycorticosterone, THDOC), dihydrodeoxycorticosterone (DHDOC) b)androstane steroids OH-androst-5-en-20-one (dehydroepiandrosterone, DHEA) and DHEA-sulphate (DHEAS).
As mentioned above, neuroactive steroids can exert a bimodal activity since some are agonist, they are known as GABA-agonist (i.e. THDOC), and other neurosteroids are antagonists, the so called GABA antagonists (i.e. P, DHEA, DHEAS). Moreover, the binding of specific neuroactive steroids (i.e. 3α-reduced neurosteroids) to GABAA receptors can lead to either inhibition or enhancement of
GABA inhibitory effects (Majewska, 1992). For this reason it is clear that neuroactive steroids play a crucial role in mediating many
33
brain functions and activities. For instance, pregenenolone sulfate and DHEAS-induced (GABA-antagonist) inhibition of GABAA
receptors may determine effects ranging from anxiety and excitability to seizure susceptibility. Moreover, DHEA and DHEAS, the most abundant circulating neuroactive steroids hormones in humans, can exert a neuroprotective action (Baulieu et al., 1998). Interestingly, a correlation between the decrease in DHEA and DHEAS concentrations, that are strictly related to age and stress conditions, and neuronal vulnerability to degeneration has been established. For instance, marked decrease of DHEA and DHEAS concentrations have been reported in patients affected by neurodegenerative diseases such as Alzheimer's disease and multi-infarct dementia (Näsman et al., 1991; Magri et al., 2000). Neuroprotection by DHEA and DHEAS was also observed in vivo, whereas the mechanisms by they act are still unknown (Baulieu et
al., 1998).
In addition, great deal of investigations have ascertained a correlation between effects of sex hormones and human cognitive functions. The most prominent examples are the effects of oestrogens and androgens on verbal fluency, the performance of spatial tasks, verbal memory tests and fine-motor skills (Hampson et
al., 1990; Phillips and Sherwin, 1992a; Phillips and Sherwin, 1992b).
There is a considerable documentation about oestrogenic influences on brain morphology and neurochemistry including the enhancement of the cholinergic system that is involved in learning and memory (McEwen et al., 1995; McEwen and Alves, 1999). All brain activities induced by neuroactive steroids are subjected to
sex-34
specific differences. Thus, oestrogen effects differ quantitatively or qualitatively between sexes.
Finally, oestrogens may exert neurotrophic and neuroprotective effects such as induction of neuritis outgrowth, dendritic spines, and synaptogenesis. They also influence the neuronal excitability, improve gene expression and exhibit intrinsic antioxidant activity (McEwen and Alves, 1999).
Fig. 10. Steroids hormones pathways.
Steroid synthesis in the brain: conversion of cholesterol to pregnenolone
The first step in the synthesis of all steroid hormones is the conversion of cholesterol to pregnenolone, catalyzed by the enzyme cytochrome P450 cholesterol side-chain cleavage enzyme (PS450scc)
(Fig. 10). In addition to adrenal glands and gonads that are the main sources of steroid production, P450scc is also present in the placenta,
35
Recently, it has been identified the localization of P450scc mRNA in
specific areas of human brain. High amounts of P450scc mRNA were
found in temporal and frontal neocortex, subcortical white matter from the temporal lobe and in the hippocampus of children and adults. Interestingly, concentrations in the temporal lobe increase markedly during childhood and reach adult levels at puberty. Moreover, P450scc mRNA levels were significantly higher in the
temporal and frontal neocortex as well as in the hippocampus of women in comparison with men (Beyenburg et al., 1999). These data establish a crucial age- and sex-dependent expression of P450scc
mRNA in the human brain and provide clear evidence that pregnenolone can be synthesized in the central nervous system.
Steroid synthesis in the brain: conversion of androgens to estrogens
Cytochrome P450 aromatase catalyzes the conversion of androgens to estrogens in specific temporal and frontal brain areas (Fig. 10). Androgens may be metabolized in the brain following two different pathways. The aromatase pathway consists in the transformation of testosterone into estradiol and androstenedione into estrone; this is similar to 5α-reductase pathway which converts testosterone into dihydrotestosterone and occurs in the majority of the peripheral androgen dependent structures (e.g. prostate).
Aromatase activity has been found higher in the placenta than in the human brain. Moreover, aromatase activity both in temporal and in frontal brain areas has been demonstrated. Interestingly, frontal aromatase activity was always higher than temporal aromatase activity regardless of sex and/or disease state. By contrast, in temporal lobe and hippocampus, the brain areas in which
5α-36
reductase was ascertained, the expression levels of 5α-reductase are very similar (Wozniak et al., 1998).
3β-hydroxysteroid dehydrogenase: inactivation of metabolite of steroid hormones
In liver metabolism, 3β-hydroxysteroid dehydrogenase (3β-HSD) plays an essential role leading to physiologically inactive metabolites of steroid hormones (Fig. 10). However, three functional 3β-HSD isozymes (type 1, 2 and 3) have been characterized on the basis of their affinity for 5α-dihydrotestosterone. The isozyme 1 is expressed exclusively in the liver, whereas the isozyme 2 is present both in the human hippocampus and liver (Khanna et al., 1995).
17β-hydroxysteroid dehydrogenase: regulation of the biological activity of steroid hormones
There are seven human isozymes of 17β-hydroxysteroid dehydrogenase (17β-HSD) (Fig. 10). 17β-HSD plays a crucial role in the regulation of the biological activity of sex hormones. 17β-HSD is essential for the biosynthesis of the strong androgens and oestrogens, (testosterone and estradiol) from their weaker precursors androstenedione and estrone (Peltoketo et al., 1999). These conversions are reversible and thus can lead to a deactivation of the respective sex hormones. The different isozymes show an individual cell-specific expression as well as a typical substrate specificity. The importance of the 17β-HSD activity in the maintenance of physiological levels of estradiol and testosterone is reflected by its ubiquitous distribution in peripheral tissues. As it regards the central nervous system, 17β-HSD expression has been ascertained not only in specific areas of the adult brain such as human temporal lobe and
37
hippocampus but also in human fetal brain. The expression levels of 17β-HSD are significantly higher in the subcortical white matter than in the cerebral neocortex. The main expression of 17β-HSD in the subcortical white matter suggests that glial cells could play a role in the biosynthesis and deactivation of sex steroids in the brain. Finally, there is no sexual dimorphism in the expression or activity of 17β-HSDs (Martel et al., 1994).
38
2-METHOXYESTRADIOL: BIOCHEMICAL FEATURES AND ITS ACTIVITY IN VITRO AND IN VIVO
General features
2-Methoxyestradiol (2ME) (Fig. 11) is an endogenous steroid metabolite of 17β–estradiol. It has been developed as a novel antitumor agent for its extreme effectiveness in several cancer lines
in vitro and in vivo. 2ME may also be active in cells of central and peripheral nervous system. Peculiar mechanisms of 2ME action make this molecule extremely attractive for the study of the pathogenesis and the treatment of several diseases occurring in the nervous system.
Biochemical and physiological properties
2ME is one of the most biologically active endogenous metabolite of 17β-estradiol (E2). E2 is metabolized to 2-hydroxyestradiol (2OH) through cytochrome P450scc by a
NADPH-dependent cytochrome P450-linked monooxygenase system. Then, 2OH is rapidly O-methylated at the 2-position by ubiquitously present catecol-O-methyltransferase to 2ME (Brueggemeier et al., 1989). 2ME binds the sex hormone-binding globulin (SHBG) in the blood and represents a fundamental step in the elimination of the potentially toxic cathecol oestrogens produced by proliferating cells (Gelbke et al., 1976; Dawling et al., 2001). 2ME plasma levels of women are higher than those found in men and they can be even more elevated during pregnancy (Berg et
al., 1992).
39
2-Methoxyestradiol affinity for oestrogen receptors
In spite of the fact that 2ME has a biochemical similarity with 17β-estradiol, other oestrogens and several their metabolites, 2ME estrogenic activity and its affinity for oestrogen receptors (ERs: ERα, ERβ, ERαβ subtypes) are still matter of debate.
Firstly, Martucci and Fishman have showed that 2ME has a low binding affinity for uterine cytosolic ERs and a very low biological activity as a conventional oestrogen. Indeed, these Authors have seen that continuous administrations of 2ME in ovariectomized rats did not exert any uterotrophic activity and did not modify LH plasma level (Martucci and Fishman, 1979). By contrast, in HUVEC (Umbilical Vein Endothelial Cells), MDA-MB-435 and MDA-MB-231 (Human Breast Carcinomas Cells) cell lines, 2ME has been probed exerting a significant binding affinity for ERα even if its biological activity was independent from ERs activation (LaVallee et al., 2002).
The above mentioned results disagree with findings provided by Banerjee and co-workers. Exposing GH3 (Rat Pituitary Tumor Cells), MCF-7 (Epithelial Tumour Cells) and MIA-Pa-Ca-2 (Pancreatic Adenocarcinoma Cells) cell lines to 2ME micromolar concentration, these Authors have demonstrated that, not only has 2ME a high binding affinity for ERs-α but also it can modulate ERs-α activation. Interestingly, 2ME seemed to exert a biphasic effect in cellular function being a stimulus for cell growth at low doses and having a reversed action at higher concentrations (Banerjee et al., 2003).
Finally, a study carried out on 2ME-exposed MCF-7 (Epithelial Tumoral Cells) cell line has supplied further data that make stronger Banerjee’s hypothesis. Indeed, a direct link between 2ME
40
effects and 2ME affinity for ERs has been clearly established. Taking into account Authors’ conclusions, the common notion about 2ME low affinity for ERs should has come from mere methodological misunderstandings. Indeed, a new methological approach has let them to find a direct link between 2ME high affinity to ERs as a real oestrogen agonist and 2ME biological activity even inducing, in this case, tumour growth (Sutherland et
al., 2005).
2-Methoxyestradiol has an antiangionetic activity: inhibition of HIF-1 alpha expression
2ME was initially considered as a direct inhibitor of angiogenesis since it strongly inhibits endothelial cell proliferation and cell migration in vitro. Moreover, oral 2ME-administration in mice has been seen reducing the neovascularization and suppressing growth of solid tumours (Fotsis et al., 1994). The mechanisms involved in such a 2ME-antiangionetic activity have been partially clarified. Studies carried out on MDA-MB-231 (Human Breast Cancer Cells), PC-3 (Human Prostatic Cancer Cells) cell lines and endothelial cells have revealed that 2ME-antiangiogenic activity is mediated through inhibition of HIF-1α expression. HIF-1α, a heterodimer of α and β subunits, is a proangiogenic transcription factor that interacts with some hypoxia response elements (HREs) and enhances the transcription of multiple proangiogenic proteins including VEGF-A (Vascular Endothelial Growth Factor). In normotoxic conditions, HIF-1α is degraded by the proteasome system, whereas it can be stabilized by hypoxia conditions (Mabjeesh et al., 2003). A recent study has presented new evidences on the role of 2ME as inhibitor of angiogenesis. It has
41
been highlighted that 2ME can determine a marked depolymerization of the microtubules in EC (Endothelial Human Pulmonary Cells) cell line inducing evident lung vascular barrier dysfunctions (Bogatcheva et al., 2007).
2-Methoxyestradiol exerts an antimitotic activity: abnormal spindle, tubulin depolymerization & alteration of tubulin dynamics
Antimitotic activity was the first mechanism that has been discovered being cause of 2ME cytotoxic effect in vitro. Since almost 20 years ago, Seegers and co-workers have probed that 2ME determined anomalies during cellular mitosis at high concentrations (between 10 and 25µM). In MCF-7 (Human Breast Adenocarcinoma Cells) and HeLa (Human Cervical Adenocarcinoma Cells) 2ME exposed-cell lines, 2ME was discovered to determine an abnormal and fragmented spindle formation (also known as bipolar) with disoriented microtubular arrangement in the metaphase of dividing cells (Seegers et al., 1989).
Abnormal polymerization of mitotic spindle was observed in SK-OV-3 (Human Ovarian Carcinoma Cells), HeLa (Human Cervical Carcinoma Cells), PC3 and DU 145 (Human Prostate Carcinomas Cells) cell lines as well. Low 2ME micromolar concentrations have been seen interfering with the normal polymerization of the mitotic spindle microtubules. Indeed, anomalous spindles appeared to be short and not properly organized so that the consequent dysfunctional alignment of DNA determined the mitotic arrest. Moreover, high 2ME micromolar concentrations exposure could trigger a marked tubulin depolymerization. In details, from 5 to
42
10µM 2ME micromolar exposure, 50% of microtubules were lost, whereas a complete inhibition of tubulin polymerization was observed between 16.6 and 19.9µM 2ME exposure (Tinley et al., 2003).
By contrast, some studies have hypothesized that 2ME exerts its antimitotic activity more intensively altering tubulin dynamics rather than determining microtubular depolymerization. In a study carried out on MCF-7 (Human Mammary Carcinoma Cells) cell line, Attalla and co-workers have reported that, the mitosis block following 2ME micromolar concentration (between 5 and 20 µM) exposure was not exclusively accompanied by depolymerization of tubulin and resultant inhibition of mitotic spindle assembly. The mere defect in mitosis was confirmed by the arrest of chromosomes in the metaphasal plate since cells whose mitosis was blocked by a tubulin depolymerizing drug (i.e. colcemide) showed their chromosomes dispersed in the cytoplasm. According to this study, 2ME mitosis-block is similarly induced by several anti-calmodulin agents or some compounds that affect microtubules dynamics (i.e. taxol and vinblastine). Thus, given the fact that metaphase to anaphase transition is a calmodulin-dependent step and 2ME inhibits calmodulin activity in vitro, 2ME metaphasal arrest has been proposed to occur via inhibition of calmodulin (Attalla et al., 1996).
Brueggemeier and co-workers., agree with the hypothesis that 2ME antimitotic effect is due to 2ME-induced alteration of microtubular dynamics rather than the solely microtubule depolymerization. These Authors have ascertained that bipolar spindle formation, block of mitosis and disruption of microtubules in MCF-7/ER+, MDA-MBA-231/ER (Breast Cancer Cells) cell lines were induced
43
starting from 8.7µM exposure (Brueggemeier et al., 2001). Recently, Kamath and co-workers have provided new evidences about such 2ME property. In details, they ascertained that in MCF7 (Human Mammary Carcinoma Cell) cell line 2ME-induced mitotic arrest was not accompanied by detectable microtubular depolymerization. The Authors found that tubulin depolymerisation 2ME-induced needed 2ME concentrations higher than those necessary to affect microtubular dynamics. These findings support the assumption that, at its lowest effective concentrations, 2ME may cause mitotic arrest and suppress microtubular dynamics by stopping microtubule growth parameters, increasing the time of steady-state and interfering with chromosome progression or cell division. This hypothesis could be also reinforced by the fact that the mitosis block 2ME-induced on actively dividing cells is more effective if they have a high proliferation rate (Kamath et al., 2006).
2-Methoxyestradiol alters cell mobility, cell adhesion and trans-well migration
2ME cytotoxic activity could be related to the alteration of other cytoskeletal functions such as cell motility, cell adhesion and trans-well migration. 2ME was established to arrest spontaneous BCR-ABL transformed Ba/F3 cells (Mouse Peripheral Blood, pro B) mobility by changing their morphology and volume. Cells exposed to micromolar 2ME concentrations (from 1 to 5µM) showed a reduction in number of pseudopodia and were detached from fibronectin-coated surfaces in which they have grown. In addition, 2ME markedly inhibited the spontaneous cell migration trough a trans-well membrane. Functional modifications were accompanied
44
by depolymerization of tubulin, appearance of disorganized microtubules and nuclear morphological features typical of defective mitosis. Taken together, these data suggest that 2ME regulates multiple cellular functions related to the main physiological cytoskeletal activities (Sattler et al., 2003).
2-Methoxyestradiol interacts with tubulin and inhibits colchicine–binding site in vitro and in vivo
Depending on the in vitro reaction conditions, 2ME can bind to the colchicine site of tubulin, be incorporated in a polymer resulting in altered stability properties or block tubulin polymerization. In presence of 1.0M glutamate/1.0mµ and MgCl2 at 37°C, 2ME
arrests the nucleation and propagation phase of tubulin assembly, whereas do not interfere with the reaction extent. Interestingly, the polymer formed in presence of 2ME is cold-stable and has little morphological abnormalities. Under sub-optimal conditions (0.8M glutamate/absence of MgCl2 at 25-30°C) 2ME could totally inhibit
tubulin polymerization (D'Amato et al., 1994).
2ME can bind to the colchicine binding-site of tubulin in unpolymerizated α-β tubulin dimers as well. In glutamate-induced tubulin assembly and in preformed tubulin, significant amounts of 2ME interact with tubulin forming abnormal polymers with altered properties (Hamel et al., 1996). The complete block of tubulin polymerization in MDA-MB-435 cells (Human Breast Cancer Cells) has confirmed the 2ME capability to bind the colchicine binding-site (Cushman et al., 1995).
45
2ME can induce apoptosis in cancer cell lines through several pathways. 2ME may determine inhibition of the transcription of superoxide dismutase (SOD) which protects cellular from damage and cell death determined by superoxide radicals. Tumour cells are more dependent on SOD than normal cells, since they have a higher superoxide production but a lower level of SOD than normal cells (Huang et al., 2003). This is one of the mechanisms by which 2ME is extremely effective in fast-growing cells (i.e. cancer cells), whereas it does not affect slow-growing cells (i.e. normal cells). When exposed to 2ME micromolar concentrations (from 1 to 2µM), normal skin fibroblast strain HSF43 (Human Fibroblast Cells) is not susceptible to 2ME.
In SV40 (Simian Vacuolating Virus 40) T antigen-transformed, HSF43 cell line, 2ME determines G2/M block, nuclear
fragmentation, micronucleation and increase in p53 peculiar of apoptosis. The Authors argued that 2ME, being a microtubule poison, it may act as enhancer of 2ME apoptotic effect. Similar results were obtained from TK6 and WTK (Lymphoblast Cells) cell lines (Seegers et al., 1997).
In addition, it has been ascertained that 2ME can start various mitogen-activated protein (MAP) kinases such as MAPKs/ERK, JNK or p38 MAPKs involved in the regulation of some cellular activities. MAPKs/ERK is a signal pathway coupling growth factors of intracellular responses to cell receptors. MAPKs/ERK is activated by mitogenic stimuli and plays a crucial role in regulation of cell proliferation and differentiation. Moreover, JNK and p38 MAPK control gene expression, mitosis and cell differentiation or survival. JNK/p38 MAPK are triggered by pro-inflammatory cytokines and environmental stresses such as osmotic shock, UV
46
light, heat shock or growth factor withdrawal. Finally, Bcl-2, a family of genes and proteins that govern mitochondrial outer membrane permeabilisation (MOMP) and may be either pro-apoptotic or anti-pro-apoptotic factors, can be phosphorylated by MAPK kinases activation. Bcl-2 phosphorylation is a crucial biochemical event related to the triggering of cellular apoptosis. 2ME through MAPK indirect or direct activation may induce Bcl-2 phosphorylation and consequent cellular apoptosis. Several studies have highlighted this assumption. In K5628 (Human Erytromyeloid Leukemia) cell line, 2ME-induced microtubular damages may determine phoshorylation of Bcl-2 uncoupling from JNK/SAPK activation. However, the Authors hypothesized that JNK/SAPK is indirectly involved in the pathway that leads to Bcl-2 activation (Attalla et al., 1998). By contrast, in OVCAR-3 (Human Ovarian Carcinoma Cell) cell lines it was found that Bcl-2 phosphorylation is solely related to p38 MAPK activation, whereas JNK or ERK are not necessary to Bcl-2 phosphorylation (Bu et al., 2006). These findings disagree with a study carried out on LNCaP (Human Prostate Cancer Cells) cell line where JNK activation was discovered to be the crucial pathway implicating in Bcl-2 phosphorylation. The data obtained let the Authors to form the hypothesis that, different mechanisms leading to Bcl-2 phosphorylation may be due to MAPKs exhibition of cell-type specific responses (Shimada et al., 2003).
2-Methoxyestradiol modifies tubulin expression
Gokmen-Polar and co-workers has compared the effects of 2ME on stable 2ME-resistant cells with a β-tubulin mutation in MDA-MB-435 (Human Breast Cancer Cells), WMDA-MB-435 (Human Carcinoma
47
Cells) and P435 (Parental Cells) cell lines. A dose-dependent depolymerization of microtubules was observed when P435 cells were exposed to 2ME micromolar concentrations, whereas 2ME-resistant cells remained unaffected. Indeed, IC50 were 0.38µM for
P435, 12.29µM for W435 and 15.23µM for 2ME-resistant cells. Interestingly, levels of acetylated and detyrosinated α-tubulin decreased in P435 cells in a clear dose-dependent manner, while they were unchanged in 2ME-resistant cells. Taken together this data suggest that 2ME induces microtubules alteration only in cells with tubulin isoforms that have affinity for 2ME (Gokmen-Polar et
48
2-METHOXYESTRADIOL-EXPOSED GLIAL AND
NEURONAL CELL LINES: LITERATURE REPORTS
2-Methoxyestradiol-exposed cells of glial origin
In a study carried out on DAOY, D341, D283 (Human Medulloblastoma Cells) cell lines and U87 and T98 (Human Astrocytoma/glioblastoma Cells) high-grade anaplastic cell lines, it has been shown that 2ME acts as inhibitor of cell growth more efficiently in tumour than in non-tumour cells. At 0.5, 1, 2 and 3µM concentrations, 2ME causes 38%, 80% decrease in growth of astrocytoma/glioblastoma and medulloblastoma cells respectively, by blocking cycle progression mainly in G2/M phase. This study also proved that 2ME-mediated growth inhibition occurs with induction of apoptosis. Interestingly, 2ME-induced apoptosis was not accompanied by modification of the expression of p53 and Bax (Kumar et al., 2003).
Moreover, another work carried out on U87, T98, U138 (Human Astrocytoma/glioblastoma Cells) cell lines and cultured rat astrocytes 2ME-exposed (from 0.001 to 10µM for 2 days) determined a marked growth inhibition (between 60% and 90%) only in glioblastoma cells. 2ME antiproliferative effect was accompanied by an increase in apoptotic cells with abnormal nuclear morphology. Treatment of U87, T98 and U138 cells with 3.3µM 2ME resulted in 39%, 20% and 82% growth inhibition in comparison with control and cells were blocked at G2/M phase. In addition, elevated levels of the tumour suppressor protein p53 were only detectable in U87 and in rat astrocytes. According to the Authors’ conclusions, given the fact that rat astrocytes have a much slower proliferation rate than glioblastoma cells and that 2ME did
49
not induce nuclear alterations in astrocytes, 2ME appears to be less effective in astrocytes than in glioblastoma cells (Lis et al., 2004). Chamaon and co-workers observed that 2ME-exposure (0.2, 2 and 20µM for 6 days) of U87, U138, LN405 (Human Glioblastoma Cells) cell lines and RG2 (Rat glioma Cell) cell line caused a significant reduction of the viable cell number. When compared with control, the drug-specific cell reduction was in the range between 10% and 40% for lower 2ME concentrations and between 75% and 100% for higher 2ME concentrations. Caspase-3 activity increase, nuclear fragmentation and micronucleation in all 2ME-treated cells, revealed the apoptotic nature of cell death. Interestingly, the Authors observed that during interphase, 2ME-induced morphological alterations were not accompanied by complete distraction of microtubular network. These findings were suggestive that 2ME had no effect on interphase microtubules since it needed a much higher concentration to generate aberrant mitotic spindles resulting in mitotic block an induction of apoptosis (Chamaon et al., 2005).
A recent study carried out on the cell lines above mentioned (U87, U138, LN405 and RG-2) confirmed that 2ME-exposure (2 and 20µM for 24 and 78 h) determined a significant reduction of the viable cells number. Moreover, the death-inducing apoptotic mechanism was further probed by micronucleation and nuclear fragmentation, whereas the role of the death receptor 5 remained controversial. Indeed, short incubation with 2ME (20µM for 24 h) determined death-cell independently of death receptor 5 up-regulation (Braeuninger et al., 2005).
50
2-Methoxyestradiol-exposed cells of neuronal origin
Retinoic acid (RA)-differentiated human neuroblastoma (SH-SY5Y) cell line was sensitive to 2ME in a concentration-dependent manner from 0.3 to 10µM 2ME micromolar exposure induced cell death and internucleosomal DNA fragmentation peculiar of defective mitosis. Interestingly, it has been found that 2ME may induce the protein synthesis inhibitors cycloheximide and apopain (Asp-Glu-Val-Asp-H, aldehyde) in inhibitor-sensitive neuronal cell death (Nakagawa-Yagi et al., 1996).
